What hardware do you need to run Nanbeige4.3 33B?

Nanbeige4.3 33B at Q4_K_M is 21 GB of weights, with a context ceiling of 512k tokens. Nanbeige built it for coding and tool use, and it is dense: every one of its 33.12B parameters runs for every token, so a token reads the whole 21 GB on disk. English and Chinese. On the cheapest machine that runs it, a Radeon AI PRO R9700, 32GB, it generates 20 tok/s at 32k of context, worked out from memory bandwidth.

How good is Nanbeige4.3 33B, really?

It has not been placed on the intelligence index yet. See the whole table.

What Nanbeige4.3 33B costs either way

Nobody rents Nanbeige4.3 33B by the token. The closest hosted match, Gemma 4 31B it, costs $0.1 per million input tokens and $0.34 per million output (OpenRouter, cheapest active endpoint, checked 2026-10-30). Buying a machine only beats that if you use it hard enough, for long enough, that the hardware price divides down below the rental bill. What a million tokens costs each way puts the two prices side by side.

Machines that run Nanbeige4.3 33B

Every machine here runs it, but not to the same length. The AMD Radeon AI PRO R9700, 32GB stops at 32k; the DGX Spark GB10 Grace Blackwell, 128GB takes it to 256k, the longest the calculator offers. The weights are the same size on every machine; what differs is the memory left for the key-value cache, which grows with every token you keep.

The table is the cheapest machine in each family, not the only one that runs it: 36 of the 56 machines on this site hold it at 32k. Within a family only memory decides, so that is every NVIDIA card with 32GB or more, every Mac Studio, Mac mini and MacBook Pro with 48GB or more, every Strix Halo box with 64GB or more, the Radeon AI PRO R9700, 32GB and the DGX Spark, 128GB. One table sets every machine against the others: price, the memory its GPU can use, what it holds and how long it takes to pay for itself.

One machine per family, cheapest first. Speeds are measured where a public benchmark exists and estimated from memory bandwidth otherwise; the calculator says which for any configuration. The longest context is the longest setting the calculator offers that the machine still holds this model at, cache included, and no machine is shown taking it past its own 512k limit. Each one opens the calculator on that machine at that length.

For scale, the calculator starts from 80 tok/s for a hosted API and times a local machine against it. Nothing above reaches it, and the quickest is 45 tok/s.

The cheapest machine that runs it is sold with 32 GB. There is a page on what 32 GB runs, machine by machine: it names the 6 machines here sold at that size, and the 33 current models the roomiest of them holds at 32k, this one among them.

Run Nanbeige4.3 33B at 256k on the DGX Spark GB10 Grace Blackwell, 128GB

How much memory does Nanbeige4.3 33B need at 256k?

Every figure above is priced at 32k of context. At 256k, the longest the calculator offers, Nanbeige4.3 33B needs 76.6 GB rather than 27.8 GB. The weights are the same either way; the key-value cache goes from 7 GB to 55.8 GB.

26 of the 37 machines here still sold at a published price hold Nanbeige4.3 33B at 32k and 12 hold it at 256k. The cheapest still sold goes from the Radeon AI PRO R9700, 32GB at $1,299, card only to the Framework Desktop, 128GB at $3,449. The Radeon AI PRO R9700, 32GB is priced as the card alone, so add the PC around it before comparing it with a complete computer.

On the Framework Desktop, 128GB it runs 6.9 tok/s estimated at 32k and 2.5 tok/s estimated at 256k. Every token generated reads the whole cache back, so a longer window is slower as well as bigger.

Two more machines, at a shorter window

The table above lists the machines that run it at 32k, the context the calculator starts on, where Nanbeige4.3 33B needs 28 GB. Two more machines hold it at a shorter window, because the weights are the same size either way and the key-value cache is not. The cheapest of those is the Framework Desktop, 32GB at $1,269. The table above starts at $1,299, card only.

One machine per family, cheapest first, from the same list as the table above. Each window is the longest setting the calculator offers that the machine still holds this model at, and it opens the calculator on that machine at that length. What it needs there is the weights plus the key-value cache at that window, measured against the memory each machine's GPU can address, which that machine's own page gives.

The specifics

Parameters

33.1B

Quantisation

Q4_K_M

Weights on disk

21 GB

KV cache

7.0 GB at 32k context — Every one of the 52 layers caches the whole context, at eight KV heads by 128 wide, so nothing here slows the cache down as it grows. How weights and cache add up.

Maximum context

512k tokens

Licence

Apache 2.0

Sources

Nanbeige/Nanbeige4.3-33B, bartowski/Nanbeige_Nanbeige4.3-33B-GGUF